Methods Glossary
This glossary accompanies the pro-regime information ecosystem classification system.
The no-code backlink analysis methodology and pro-regime information classification ecosystem accompany this glossary.
Appearance
An appearance is an instance of a single URL appearing in the data. The Appearance may be unique or a repeat. If there are URLs to five articles titled Article 1, Article 2, and so on, hosted on the domain example.com, then the domain example.com will be listed as appearing five times for domain level data. The URLs will be listed as appearing once each.
In the example image, we see the domain is hoaxlines.org. URLs are the four specific pages (The Basics, Ukraine Dashboard, Information Classification, and Ukraine Dashboard again).
The Basics page: https://hoaxlines.org/mis-dis-mal
Ukraine Dashboard: https://hoaxlines.org/ukraine-dashboard
Information Classification: https://hoaxlines.org/info-classification
Ukraine Dashboard: https://hoaxlines.org/ukraine-dashboard
If we counted the number of URLs or appearances for Hoaxlines, we would get four. If we counted each unique URL appearance, we would count three unique URLs. Let us pretend these four URLs in the example are in our backlink study because each of them had backlinked Example.com (our pretend core website). Hoaxlines.org would be listed as appearing in one core website dataset, that of Example.com. The number of appearances for Hoaxlines.org across the entire study—pretend there are other datasets besides Example.com— would be four, unless Hoaxlines.org appeared in another dataset. If it does not, then we have four appearances for the domain Hoaxlines.org in one core dataset.
The arrows pointing to the “core website” represent hyperlinks on these pages that lead to the core website, Example.com. We call these backlinks. Techopedia defines a backlink as “a hyperlink that is used to link one website (that is adding the link) with another website (whose link is added and is receiving the backlink).”
Based upon the second of the two diagrams, if you go to “The Basics” page, you should see a link to the Core website, Example.com, somewhere on that page. The link may be hidden as a hyperlink, which is when a link has been embedded in a text like this. If you clicked “like this,” you would find out that Hoaxlines.org is hyperlinked to the text.
For the Information Ecosystem Mapping and Classification, two values are calculated for each domain. One reflects the number of core website datasets each domain appeared in, and the other is the number of total appearances for that URL overall across all datasets. In the case of websites with more than 10,000 backlinking URLs in a year, like Zero Hedge and RT, Buzzsumo provided a sample of 10,000 URLs.
Core channel or website
The term “core” reflects that these are the domains with which we began our network analysis. It does not necessarily suggest the closeness to a specific regime, but does indicate that the website has published pro-regime content. See pro-regime definition. Our trial of this method included 42 core websites, but that number will vary based upon the regime or ideology being mapped.
Domain
A website address that generally contains the name or abbreviation of the website. If a URL is a room in a house, then the domain is the house. The domain of the example URL is hoaxlines.org.
Interquartile Range (IQR)
The interquartile range (IQR) is the gap between the upper quartile (Q3) and lower quartile (Q1). It shows the middle 50% of values when they are listed from lowest to highest. The IQR is usually seen as a better way to measure range than the regular range because it is not impacted by outliers.
Pro-regime
We define pro-regime as framing, arguments, or ideas endorsed, promoted, or implied by regime officials, employees, representatives, allies, or contract workers. The pro-regime label does not indicate whether something is true, false, misleading, or otherwise.
Therefore, we define pro-Kremlin as framing, arguments, or ideas endorsed, promoted, or implied by Kremlin officials, employees, representatives, allies, or contract workers. The pro-Kremlin label does not indicate whether something is true, false, misleading, or otherwise.
Pro-regime website
Whether pro-regime is positive or negative is in the eye of the beholder. The pro-regime label does not indicate whether something is true, false, misleading, or otherwise. The only trait inclusion in the pro-regime information system indicates is that the site met Hoaxlines’ criteria for a pro-regime website.
Outlets are classified as Tier 1, Tier 2, Tier 3, Interface, or Multi-State.
While some of these outlets have concrete ties to the Russian state, some included in the data do not.
Inclusion in the database should not be regarded as an assertion that an outlet is affiliated with the Russian state nor that the story is false, misleading, or otherwise nefarious.
Entities included in this information ecosystem report met one of three criteria.
The website is recognized by the US federal government or an intelligence agency from one of the Five Eyes (Canada, Australia, New Zealand, the United Kingdom, and the United States) as pro-regime or a regime agent.
The website is recognized as pro-regime or another comparable label by a research organization or media manipulation specialty outlet that employs qualified scientists, researchers, ethical investigative journalists, or other appropriate experts and shares the evidence supporting their conclusions.
Outlets or organizations engaging in any one of the following:
Repeatedly publishing, promoting, or citing pro-regime content or authors from an outlet that meets criteria one or two,
A Tier 1 organization or representative owns another outlet’s domain, oversees editorial decisions, or provides funding to a website.
The website owner has previously produced content for the Russian state through either overt propaganda like RT or covert, such as the outlets controlled by the GRU or FSB. And the website has published recognized disinformation that can be clearly tied to Russia, as with the 2017 Ukraine “biolabs” story published by GRU front, Cyber Berkut. Russian and Chinese state-affiliated accounts boosted the story on social media.
A website publicly shared a partnership with a Tier 1 or Tier 2 entity, as demonstrated by disclosure on at least one of the partner sites that goes legally uncontested, public records, reporting from reliable news outlets, public statements, or other behavior that suggests intentional collaboration. Anything a disinterested party may reasonably argue is a coincidence does not fulfill this requirement.
Repeatedly
The term “repeatedly,” as used in the tier classification system, is defined as three or more instances in two years, regardless of if the content is withdrawn or removed from the site. The time frame reflects the significant time taken by some to produce content.
Share Count Group
A group of URLs or Unique Domains with the same number of appearances or shares across all datasets.
Target
Links collected from the core Telegram channels are called “targets” or “links.” Although these could also be called “sources,” we refrain from using source unless talking about a core channel.
URL
A URL (uniform resource locator) is a complete web address to a specific page, like an article or an “about us” page. An example of a URL is https://hoaxlines.org/mis-dis-mal. See domain for comparison and appearance for an illustrated example.
URL share count
A URL share count reflects the number of URLs. If the URL share count appears in a table related to core channels, then the URL share count indicates the number of URLs shared a channel. If it appears in a table looking at targets, the links collected from each of the core channels, then the URL share count reflects the number of URLs from the listed target.